ShijieZhou-UCLA / feature-3dgs

[CVPR 2024 Highlight] Feature 3DGS: Supercharging 3D Gaussian Splatting to Enable Distilled Feature Fields
Other
374 stars 23 forks source link

IndexError: list index out of range for render.py with sam and speedup #10

Closed cs-mshah closed 8 months ago

cs-mshah commented 8 months ago

I'm facing the following error with -f sam --speedup:

python render.py -s data/360_v2/room -m output/360_v2/room/4 -f sam --iteration 7000 --novel_view --speedup
Looking for config file in output/360_v2/room/4/cfg_args
Config file found: output/360_v2/room/4/cfg_args
Rendering output/360_v2/room/4
Loading trained model at iteration 7000 [27/03 17:34:38]
Reading camera 311/311 [27/03 17:34:39]
Loading Training Cameras [27/03 17:34:39]
Loading Test Cameras [27/03 17:35:20]
Rendering progress: 100%|___________________________________________________________________________________________________________________________________| 311/311 [01:06<00:00,  4.69it/s]
Traceback (most recent call last):
  File "/data/manan/feature-3dgs/render.py", line 389, in <module>
    render_sets(model.extract(args), args.iteration, pipeline.extract(args), args.skip_train, args.skip_test, args.novel_view, 
  File "/data/manan/feature-3dgs/render.py", line 356, in render_sets
    render_set(dataset.model_path, "test", scene.loaded_iter, scene.getTestCameras(), gaussians, pipeline, background, edit_config, dataset.speedup)
  File "/data/manan/feature-3dgs/render.py", line 114, in render_set
    gt_feature_map = views[0].semantic_feature.cuda()
IndexError: list index out of range

same issue comes without novel_view flag. I've already finished the training using:

python train.py -s data/360_v2/room -m output/360_v2/room/4/ -f sam -r 0 --iterations 7000 --speedup
cs-mshah commented 8 months ago

Fixed: I added --skip_test flag.